A narrow band finite element method for the level set equation

Maxim Olshanskii,Arnold Reusken,Paul Schwering
2024-07-03
Abstract:A finite element method is introduced to track interface evolution governed by the level set equation. The method solves for the level set indicator function in a narrow band around the interface. An extension procedure, which is essential for a narrow band level set method, is introduced based on a finite element $L^2$- or $H^1$-projection combined with the ghost-penalty method. This procedure is formulated as a linear variational problem in a narrow band around the surface, making it computationally efficient and suitable for rigorous error analysis. The extension method is combined with a discontinuous Galerkin space discretization and a BDF time-stepping scheme. The paper analyzes the stability and accuracy of the extension procedure and evaluates the performance of the resulting narrow band finite element method for the level set equation through numerical experiments.
Numerical Analysis,Computational Physics
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to develop an efficient and accurate narrow - band finite element method for solving the level - set equation. Specifically, the author aims to improve the extension process in the existing narrow - band level - set methods by introducing an extended method based on the finite element \(L^2\) or \(H^1\) projection combined with the ghost - penalty technique. This method can ensure computational efficiency and is suitable for strict error analysis. ### Main problems and solutions 1. **Problem background**: - The level - set method is widely used to represent and approximate moving interfaces (or surfaces), especially in fields such as fluid mechanics and materials science. - The narrow - band level - set method improves computational efficiency by solving the level - set equation within a narrow region near the interface. However, these methods require effective extension techniques to handle the initialization of new regions in the time - stepping process. 2. **Limitations of existing methods**: - Traditional re - initialization techniques such as the fast marching method (FMM) are effective but have difficulties in controlling the change of surface positions. - Other methods such as solving elliptic PDEs also have problems such as complex parameter selection and high computational cost. - These methods lack strict error estimates and are difficult to be combined with finite element analysis. 3. **Solutions proposed in the paper**: - A new extension method is introduced, which is based on the finite element \(L^2\) or \(H^1\) projection combined with the ghost - penalty technique. - The ghost - penalty technique enhances the stability of the finite element formulation and is especially suitable for the Eulerian finite element formulation in time - dependent domains. - The extension method is formulated as a linear variational problem, which is carried out within a narrow region near the interface, making it computationally efficient and easy to conduct strict error analysis. 4. **Specific implementation**: - The level - set equation is solved using the discontinuous Galerkin (DG) space discretization method and the BDF time - stepping scheme. - Combined with the newly proposed extension method, it is ensured that the new initial values after each time step are accurate and stable. 5. **Theoretical analysis and numerical experiments**: - The stability and accuracy of the extension method are analyzed, and its performance is verified through numerical experiments. - The results show that the new method not only has a good error estimate in theory but also shows excellent performance in practical applications. ### Summary The main contribution of this paper is to propose a new extension method, which solves the key problem in the narrow - band level - set method, that is, how to ensure high accuracy and stability while maintaining computational efficiency. This method provides a powerful tool for research in related fields and has important application value especially when dealing with complex interface evolution problems.